1. Field
One or more embodiments of the present disclosure relate to a three-dimensional (3D) image generation apparatus and method that may generate a 3D image from a two-dimensional (2D) image and a depth image, and more particularly, to a 3D image generation apparatus and method using a region extension of an object in a depth map.
2. Description of the Related Art
When a three-dimensional (3D) virtual image is generated from a two-dimensional (2D) color image and a depth image, the depth image may be inaccurate when noisy or when the depth image is automatically restored using computer graphics technology. When a 3D virtual image is generated based on the depth image, objects included in an original image may not preserve a boundary or a shape of each of the objects in the 3D virtual image.
When viewing a 3D virtual image generated as described above, a stereoscopic effect may be reduced or a user may feel more fatigued, since a boundary of an object may not be maintained or a shape of the object may be changed.
Accordingly, a technology that can preserve a shape of an object included in an original image and generate a high-quality 3D virtual image when a 3D virtual image is generated from a color image and a depth image is required.